Posts by Posts on R Lover ! a programmer
Author: Posts on R Lover ! a programmer
Warspeed 5 — priors and models continued
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. My last 4 posts have all focused on thevaccines being produced to fight COVID-19. They have primarily focused onBayesian methods (or at least comparing bayesian to frequentist methods). Thisone follows that pattern and provides expanded coverage of the concept ofpriors in bayesian thinking, how to operationalize them, and additionalcoverage of how to compare bayesian regression models using various tools inr. There’s no actual additional analysis of “real data”. While the news has allbeen good lately I haven’t found anything publicly available that begs forinvestigation. Instead we’ll use “realistic” but not ... Read More
Exploring vaccine effectiveness through bayesian regression — Part 4
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. Yet another post about COVID-19 vaccines. Ironically I started with Modernafor the first two poststhen shifted to Pfizer for the 3rd postwhen they released data first. Yesterday Moderna actual data became available (greatnews), along with some additional information we can model. Literally as I waswriting the post more Pfizer data came out with even more good news. Last time Last time we paid homage toDoing Bayesian Data Analysis by Dr. John K. Kruschke (hereafter, simply DBDA) while exploringoverall effectiveness of the Pfizer vaccine. This time we’ll revisit brms and STANwhich we ... Read More
Warpspeed vaccine vindication and an homage — Part 3
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. Yahoo (not the service provider), yippee, and all those other cries of joy. Aftermy last couple of posts what appears this morning… Earlier on Monday, pharmaceutical giant Pfizer and its partner BioNTech said their experimental vaccine was 90% effective in preventing COVID-19. It is the first of the vaccine attempts to reach the clinical milestone. … That news comes from an interim analysis of a study involving 43,538 volunteers, 42% of whom had “diverse backgrounds.” … The vaccine proved to be more than 90% effective in the first 94 subjects ... Read More
Warpspeed confidence what is credible? — Part 2
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. In my last post I started exploring a basic statistical analysis of howconfident we should feel about the effectiveness of a vaccine for COVID-19(Operation Warpspeed). A reminder I’m not an epidemiologist, I don’t have anydata for safety, and we’re still quite early in the Phase III trials – justapproaching the point at which they can take a first “peek” at the data (whichis doubleblind).You’ll be best served if you’ve already readthe previous post. The three key conditions that have to be (there are certainly other conditions as well): 53 confirmed ... Read More
Warpspeed confidence what is credible?
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. Well, it’s election day here in America and the last thing in the world I wantto talk about is politics. Hope you found a way to vote. Meanwhile, while wewait let me (hopefully) amuse and perhaps educate you with some informationabout another topic that we all care deeply about. A vaccine for COVID-19. The other day I was in my car listening to the radio when I heard Olivier Knoxinterviewing Dr. Moncef Slaouiabout progress towards developing a vaccine for COVID-19. Since I was driving itwas difficult to pay attention to exact ... Read More
Understanding COVID19 in Connecticut. It takes a town
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. Well, it has been way too long since I posted something. As with most of youlife during the pandemic has been interesting to say the least. I’ve beenworking hard (to include trying to find a new job), practicing good hygiene,socially distancing, wearing a mask etc., etc. But so far I have avoided thetemptation to post about the virus (or should I say the data surrounding thevirus) until now. As you may have divined I have succumbed to the trend asof today. Hopefully with good reason and good effect! From my ... Read More
Updating to 4.0.0 on MacOS
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. Mixed emotions Wow! Has it been a year? Another major update from The RFoundation (the recent 4.0.0release in April). I’m always happy to see thecontinuing progress and the combination of new features and bug fixes, but Ialso dread the upgrade because it means I have to address the issue of what todo about the burgeoning number of packages (libraries) I have installed. I wrotea fairly comprehensive post about it last year. I just took the plunge this yearand almost everything seems to still work. Vindication! The details are here in ... Read More
Analyzing churn with chaid
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. This post tries to accomplish several things concisely. I’m making available anew function (chaid_table()) inside my own littleCGPfunctions package,reviewing some graphing options and revisiting our old friend CHAID – ChiSquared (chi^2) Automated Interaction Detection – to look at modeling a “realworld” business problem.It’s based on a blog postfrom Learning Machines and investigates customerchurn for a wireless provider. The original blog post does a nice job ofdescribing a package called rattle and how it was used. I’m going to useRStudio with a variety of packages that I’ll mention along the way.The ... Read More
I didn’t mean() to ignore the median()
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. This week’s post follows directly from last week’sinvestigationof data from the 2016 US Census Bureau’s American Community Survey (ACS) PublicUse Microdata Sample(PUMS). We exploredmean differences in income across several different types of employment status(self-employed, private sector, government, etc.). We found, using bayesianmethods, strong evidence for differences across the categories and were able toplot them in a variety of formats using ggplot2 and ggsignif.While I was very happy with the plot results (which was my primary interest)something was nagging me the whole time I was writing the post. Our outcomemeasure was ... Read More
Plotting Bayes Factors for multiple comparisons using ggsignif
Feed: R-bloggers. Author: Posts on R Lover ! a programmer. This week my post is relatively short and very focused. What makes it interesting(at least to me) is whether it will be seen as a useful “bridge” betweenfrequentist methods and bayesian methods or as an abomination to both! There’ssome reasonably decent code and explanation in this post but before I spend muchmore time on the functionality I definitely want to hear some feedback. For the small number of you who have been following along I have been trying toelevate my own use of bayesian methods as well as share what ... Read More
Recent Comments